NVIDIA Tesla K40st vs NVIDIA GRID A100A
We compared two Professional market GPUs: 12GB VRAM Tesla K40st and 48GB VRAM GRID A100A to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA Tesla K40st 's Advantages
Lower TDP (245W vs 400W)
NVIDIA GRID A100A 's Advantages
Released 6 years and 6 months late
Boost Clock1005MHz
More VRAM (48GB vs 12GB)
Larger VRAM bandwidth (1866GB/s vs 288.4GB/s)
4032 additional rendering cores
Score
Benchmark
FP32 (float)
Tesla K40st
3.312 TFLOPS
GRID A100A
+319%
13.89 TFLOPS
Graphics Card
Nov 2013
Release Date
May 2020
Tesla
Generation
GRID
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
-
Base Clock
900 MHz
-
Boost Clock
1005 MHz
1502 MHz
Memory Clock
1215 MHz
Memory
12GB
Memory Size
48GB
GDDR5
Memory Type
HBM2e
384bit
Memory Bus
6144bit
288.4GB/s
Bandwidth
1866GB/s
Render Config
-
-
-
-
SM Count
108
2880
Shading Units
6912
240
TMUs
432
48
ROPs
192
-
Tensor Cores
432
-
-
-
16 KB (per SMX)
L1 Cache
192 KB (per SM)
1536 KB
L2 Cache
48 MB
-
-
-
Theoretical Performance
34.50 GPixel/s
Pixel Rate
193.0 GPixel/s
138.0 GTexel/s
Texture Rate
434.2 GTexel/s
-
FP16 (half)
55.57 TFLOPS
3.312 TFLOPS
FP32 (float)
13.89 TFLOPS
1104 GFLOPS
FP64 (double)
6.947 TFLOPS
Board Design
245W
TDP
400W
550 W
Suggested PSU
800 W
No outputs
Outputs
No outputs
-
Power Connectors
None
Graphics Processor
GK110B
GPU Name
GA100
-
-
-
Kepler
Architecture
Ampere
TSMC
Foundry
TSMC
28 nm
Process Size
7 nm
7.08 billion
Transistors
54.2 billion
561 mm²
Die Size
826 mm²
Graphics Features
12 (11_1)
DirectX
N/A
4.6
OpenGL
N/A
3.0
OpenCL
3.0
1.1
Vulkan
N/A
3.5
CUDA
8.0
5.1
Shader Model
N/A
